1. Field of the Invention
The present invention relates to a method of manufacturing a magnetic head for use in writing information into a magnetic recording medium and reading information from the magnetic recording medium. In particular, the present invention relates to a series of manufacturing steps which join a core chip blank together with a slider blank to form a magnetic head. The invention also includes a slider for use in the series of manufacturing steps and a mold for molding the slider blank.
2. Description of the Related Art
In recent years information has spread widely regarding recording/reproducing systems that use magnetic recording media. One of these systems is a floppy disk drive device that stores information into a disk-like magnetic recording medium. This disk-like magnetic recording medium is commonly referred to as a floppy disk.
Currently, typical floppy disk drives are of the 3.5-inch 135TPI (Track Per Inch) type. The magnetic heads they employ are mainly of the tunnel erase and bulk types. This practice is expected to continue into the future. The floppy disks for use in these floppy disk drives typically store two megabytes of data.
There is a trend wherein pieces of electronic equipment are becoming smaller. In order to fit inside miniaturized pieces of electronic equipment, floppy disk drive devices must also become smaller in size. Naturally this will lead to a demand for miniaturized floppy disk drive components, including the magnetic head.
Ordinarily, the magnetic head also known as a bulk-type head, comprises a ceramic slider placed in sliding-contact with the recording surface of the magnetic recording medium and a gimbal spring for supporting this slider on a carriage arm. This magnetic core accommodates a recording/reproducing coil assembly and an erasing coil assembly which are attached to the magnetic core. This magnetic core also accommodates a back core for magnetically closing the opening in the end of the magnetic core.
FIG. 1 is a schematic view of such a conventional magnetic core. FIG. 1 shows a magnetic core 104 of a type in which a core chip 101 is clamped by a first slider 102 and a second slider 103. The slider for this type of magnetic core is partitioned into two parts, and is therefore called a two-part slider. FIG. 2 depicts the magnetic core 104 shown in FIG. 1, viewed from below. In order to manufacture the magnetic core 104 of such two-part sliders, the first and second sliders 102, 103 are individually skived from a large block of ceramic with a diamond wheel. The core 101 is then clamped between the first and second sliders 102, 103 and they are all joined together. Alternative methods may include the steps of individually preparing the two sliders 102, 103 through powder compressive molding or injection molding, clamping and joining the magnetic core 101 by them, and then carrying out the finish processing. Both methods are problematic because the processing cost is high.
In the magnetic core having the two-part slider structure the core chip 101 is clamped between the first slider 102 and the second slider 103 and joined to them by adhesive or glass. The position of each of the joint surfaces may inconveniently with the result that the bottom surfaces of the two sliders 102 and 103 and core chip 101 do not lie on the same plane. That is, the core chip bottom surface 105, the first slider bottom surface 106 and the second slider bottom surface 107 shown in FIG. 2 may not coincide in the same plane but instead be uneven. FIG. 3 depicts a dispersion in distance (hereinafter referred to as CS level difference) between the plane defined by the first slider bottom surface 106 and the second slider bottom surface 107, and the core chip bottom surface 105. From FIG. 3, the width of dispersion in CS level difference is seen to be about 20 .mu.m.
Shown in FIG. 2 is a recording/reproducing gap depth 108 (gap depth is hereinafter referred to as GD). This recording/reproducing GD and the erasing GD (not shown in FIG. 2) together determine the "good" or "poor" electromagnetic conversion characteristics. This relationship is well known. In order to confer predetermined dimensions on GD, the top surface on the sliding-contact surface side is ground. This method of grinding comprises the steps of aligning several hundreds of the cores on the GD processing plate, adhering the cores to the plate, and then processing them all at one time.
At that time, in order to process so as to reduce the dispersion in GD of a group of several hundred cores 104, which are processed as a unit, a) the distance from the bottom surface 105 of the core chip to the lower end of the gap must coincide for all the core chips, and b) the dispersion in CS level difference must be small, and c) each core must be adhered to the GD processing plate in such a manner that the adhesive has a uniform thickness.
The maximum allowable dispersion in GD dimension for magnetic heads used in floppy disk drives designed to handle two megabyte floppy disks is .+-.10 .mu.m. Thus, if the dispersion in CS level difference were 20 .mu.m in width, it would be the same as the allowable dispersion in GD dimension. For that reason, the core chip 101 must be grouped in incremental depths of 4 to 5 .mu.m with respect to CS level difference, and the process must use abrasive suitable for each group. In other words, stratified operation on groups of cores is required.
With the aim of lowering the production cost, a magnetic core 110 having an integral slider as shown in FIG. 4 was conceived.
The slider 111 is provided with the two sliding-contact surfaces coming into sliding-contact with the floppy disk. The sliding-contact surface provided with an elongated hole 112 is a main sliding-contact surface 113, and the other sliding-contact surface is a subsidiary sliding-contact surface 114. The elongated hole 112 accommodates a core chip 115 which is adhered to the slider 111 by glass.
The magnetic core 110 shown in FIG. 4 is obtained by assembling the individual parts shown in FIG. 5. That is, the magnetic core 110 comprises a core chip blank 119 having an elongated hole 118, a core chip blank 120 to be accommodated into the elongated hole 118, and a glass rod 121 to be melted to join the slider blank 119 and the core chip blank 120 together. The slider blank 110 includes along the long sides of its elongated hole 119 a series of protrusions 122. The protrusions 122 are used to position the glass rod 121 when mounting the glass rod 121 on the core chip blank 119.
As described above, the glass rod is melted to allow the glass to enter the gap between the core chip blank 120 and the slider blank 111. The protrusions 122 stably mount the long glass rod 121 prior to melting. The protrusions 122 are slightly longer than the elongated hole 118.
FIG. 6 is a sectional view showing the A--A section when the individual parts in FIG. 5 are combined. Alphabetic Symbols designate dimensions of the ingredient components. Dimension a designates an inner dimension of the protrusions 122. Dimension b designates a distance between the long sides of the elongated hole 118 Dimension c denotes a diameter of the glass rod 121 Dimension d denotes a thickness of the core chip blank 120 Dimension e denotes a height of the protrusions 122 Dimension f signifies a step in lower surfaces, namely, a CS level difference between the core chip blank 120 and the slider blank 119.
The core chip blank 120 is higher than the core chip 115 shown in FIG. 4 by about 100 .mu.m, and the glass rod 121 is mounted between the protrusions 122 for melting. When the glass rod 121 is mounted prior to melting the height of the mounting surface 123 on which the glass rod 121 is mounted is greater than that of the core chip blank 120 by about 100 .mu.m. Between the core chip blank 120 an the slider blank 119 there is a clearance on the order of 20 .mu.m so that the core chip blank 120 is correctly positioned and the core chip blank 120 can be inserted into the elongated hole 118 without influence of frictional force.
The steps of manufacturing the magnetic head having the integral slider mentioned above will now be described. The slider blank 119 is placed on an assembly jig (not shown) with its sliding part upward, and the core chip blank 120 is inserted into the elongated hole from above so as to bring its bottom surface into intimate contact with the jig surface. The glass rod 121 is mounted between the protrusions 122 and then melted in an electric furnace. The glass infiltrates and fills the gap between the core chip blank 120 and the elongated hole 118, to join the core chip blank 120 to the slider blank 119.
After joining in this manner, the protrusions 122 are ground for removal together with the glass hardened after melting to skive a sliding surface. The sliding surface is further ground to obtain a smoothly finished surface, thus obtaining a desired accuracy in the GD. The subsequent steps are not important in this case, and hence figures thereof are omitted from this discussion. The coil assembly for recording/reproducing and for erasing is inserted into the core leg of the core chip. The back core is then assembled into the end of the core leg to complete the magnetic circuit. Finally, it is adhered to the gimbal spring and connected to the coil terminals to complete the magnetic head.
In the method of manufacturing the two-part slider magnetic head according to the prior art technique described above, a stratified operation on groups of cores must be used for the GD processing due to a larger dispersion in CS level difference.
Despite the fact that in construction of the integral slider magnetic head there is an absence of frictional resistance when the core chip blank is incorporated into the slider, the dispersion in CS level difference is greater than that of the two-part slider described above. Accordingly, a stratified operation on groups of cores is required at the time of GD processing, in the same manner as in the two-part slider.
As a result, in addition to increased complexity of manufacturing operations, products belonging to the stratified individual groups must be accumulated in stock until there are enough to process as a unit, since the core located at the ends of the width of the dispersion level difference is naturally of rare occurrence.
Steps even larger than the expected width of the dispersion in CS level difference sometimes occur. Since these must be considered "poor" the yield will be lowered, thus increasing cost.
When the stratified operation for CS level difference is executed, the bottom surfaces of the core chip and the slider may be subjected to stress which may cause a flaw or cutout to occur. Regardless of whether a flaw or cutout occurs, the residual set may lower the quality. Also, the offset of the glass rod may cause the glass to deficiently infiltrate and fill the gap between the core chip blank 120 and the elongated hole 118.
Since the conventional slider shrinks 20% when the slider blank 119 is sintered, it can be readily deformed. In the erect posture with the main and subsidiary sliding-contact surfaces 124, 125 facing upward as shown in FIG. 7, for example, the shrinkage is restricted by the frictional resistance of the slider blank 119 in contact with the sintering jig 126 whereby the slider presents a trapezoidal shape as shown in FIG. 8. When the slider blank 119 is laid down on its side as shown in FIG. 9, the end walls 127 may be caved inward by the weight of the slider blank as shown in FIG. 10. In the inverted state with the main and subsidiary sliding-contact surfaces 124, 125 facing downward as shown in FIG. 11, the slider is minimally deformed. In this posture, however, the stability is poor due to the presence of the protrusions 122, and this will inevitably lead to deformation. Note that the slider blank 119 shown in FIGS. 8 and 10 is extremely deformed for convenience of explanation.
Irrespective of posture taken, deformation occurs and causes the peripheral assembly reference surfaces 128 and 129 to become wavy as shown in FIG. 12. This waviness makes it difficult to correctly position the read/write head with a precise recording/reproducing gap 130 and azimuth 131. As a result, it is also difficult to register the recording/reproducing track center 131 using the outer peripheral reference surfaces 128 and 129.
Upper and lower mating surfaces of the mold for molding the integral slider blank 119 coincide with the lower end portion 132 of the elongated hole shown in FIG. 6. Therefore, a burr produced in this region will catch when the core chip blank 120 is inserted into the elongated hole, which will result in dispersed GD dimensions. In order to remove this burr, barrel processing must be done. However, the lower end portion 132 of the elongated hole is located inside the slider blank 119, and hence media for the barrel finishing do not reach there. The provision of the butt surface within the interior of the elongated hole 118 would cause an offset at this area, which would catch the core chip blank 120 at the time of insertion thereof or prevent it from being inserted.
It is therefore the object of the present invention to solve the above problems by providing a method of manufacturing a magnetic head less dispersion in CS level difference, thus a) reducing the number of GD processing steps and b) reducing the number of parts which must be accumulated in stock to a minimum and c) shortening the processing procedure while d) simultaneously ensuring a consistently high quality and a reduced production cost.
Another object of the present invention is to provide a slider for a combined magnetic head whose blank which, when sintered, remains stable in posture and minimally deforms.
A further object of the present invention is to provide a slider for a combined magnetic head which may be accurately assembled through use of an assembly reference surface.
A still further object of the present invention is to provide a mold for molding a slider for use in a combined magnetic head in which a transverse burr can project from the upper end surface of the core slit molding part. This facilitates removal of the burr and prevents an extensive dispersion in the CS level difference.